Rich technical directions and model types
 EasyDL image
EasyDL image Wenxin big model
image classification Object detection image segmentation
Based on Wenxin large model, it realizes image content understanding and classification, object detection and positioning in the image, and is suitable for image content retrieval, security monitoring, industrial quality inspection and other scenarios, and supports model customization of less labeled sample scenarios.
 EasyDL text
EasyDL text Wenxin big model
Categorize tasks Match Task Sequence annotation task
Based on the Wenxin model, it provides zero threshold customization and application capabilities of various NLP models, widely applicable to various text processing and analysis scenarios such as consultation classification, content review, content production, etc
 EasyDL voice
EasyDL voice
speech recognition Sound classification
Customized speech recognition model, accurate identification of business proper nouns, applicable to data acquisition and entry, voice instructions, call center and other scenarios, as well as customized voice classification model, used to distinguish different voice categories
 EasyDL OCR
EasyDL OCR
Character recognition
Customized training character recognition model, structured output of key field content, meet the needs of personalized card, certificate and bill recognition, and apply to electronic approval of certificates, electronic financial and tax reimbursement and other scenarios
 EasyDL video
EasyDL video
target tracking Video classification
Customized analysis of video clip content, tracking of specific target objects in the video, applicable to video content review, people/vehicle flow statistics, livestock movement trajectory analysis in farms and other scenes
 EasyDL structured data
EasyDL structured data
Table Forecast Time series prediction
Mining hidden patterns in data to solve problems such as binary classification, multi classification and regression, which is suitable for scenarios such as customer churn prediction, fraud detection and price prediction
 EasyDL Transmodal
EasyDL Transmodal Wenxin big model
Image text matching
Customize the image text matching model based on the Wenxin large model. Fusion of language, vision and other multimodal information, output the matching degree of image and text content, suitable for content quality evaluation, image and text material recommendation and other scenarios
 EasyDL Retail Industry Edition
EasyDL Retail Industry Edition
Commodity inspection Statistics of the proportion of layout Shelf splicing Display layer identification Display scene identification Reprint identification
Customized commodity detection model, based on Baidu's large-scale retail data pre training model and data enhancement synthesis technology, provides a variety of optional service functions and mature AI capabilities to fully meet the business needs related to commodity identification in the retail FMCG industry
One stop model customization
 One stop model customization
Built in Wenxin large model base

Large model of industrial level knowledge enhancement

The preferred base model for AI development

Learn more Introduction
Leading technology
Full scene coverage
Zero threshold use
Truly touchable
Comprehensive edge deployment capability

Cloud edge full scene coverage 30+chip wide adaptation

Special optimization accelerates ultimate reasoning performance

View model algorithm reasoning performance
Diversified use
Online use
Use EasyDL in the cloud to complete one-stop model training and deployment, and get started immediately without relying on the local computing environment. The model produced by the cloud can flexibly support online and offline multiple deployment schemes
Local use
The fly propeller EasyDL desktop version is installed locally, which can realize the one-stop model training and deployment process offline locally, and the data and output models can be efficiently managed locally
Selected Scenarios Example
Industrial manufacturing
smart city
Electronic Commerce
energy
retail

Automatic Identification of Scratches on Industrial Parts

By training the customized AI model of object detection, the intelligent recognition of "scratches on industrial parts" is realized, which is widely used in industrial quality inspection, intelligent manufacturing and other fields. Label: object detection, part scratch, industrial manufacturing

Experience AI
AI official account
Follow AI official account
Get AI information and experience AI ability
Back to top